The inside surface of the fingertip has intricate contours that are unique to each individual. This has made the fingerprint pattern an important vehicle for identification, particularly for law enforcement purposes. To facilitate identification via fingerprints, efforts have been made to develop fingerprint detection devices capable of electronically mapping an individual fingerprint pattern. Accurately mapping the contour of each person's fingertip, however, has proven to be a difficult and expensive task. This is because each fingerprint pattern is made up of a large number of convex portions, or ridges, separated by concave portions or grooves that generally extend in random directions around the fingertip. These ridges and grooves have minute dimensions, on the order of 100 .mu.m, making them difficult to detect.
Conventional fingerprint detection devices generally employ a surface pressure input panel that includes a contact sheet which changes in some manner when it is engaged by the various ridges and grooves of the fingertip. One such device comprises a conductive rubber sheet that undergoes local variations in conductivity when engaged by a surface with a pressure differential. Similar devices include a resistive film that varies in resistance with changes in surface pressure. These devices employ a means for detecting the changes in the contact sheet or film in order to electronically map the contour of the fingertip. The detection means, however, typically has difficulty detecting small pressure changes, particularly when a non-uniform pressing force is applied or the fingertip is contaminated with grease, sweat or the like. Therefore, these devices are often incapable of accurately detecting the small surface variations necessary for mapping the contour of a fingertip.
Other fingerprint detection devices have attempted to overcome this problem by employing MOS field-effect transistors or piezoelectric thin films to detect small surface variations (e.g., see Japanese Patent Application No. Hei 5-61966). These devices, however, generally require complicated manufacturing techniques and relatively expensive materials, such as silicon semiconductor substrates. Although these devices can produce high definition, they are difficult to manufacture and, therefore, are very expensive. Accordingly, manufacturing these devices to map a large detection area, such as a fingerprint pattern (which is very large compared to the size of the individual surface changes or irregularities) is not presently feasible.